We have focused on 4 related areas. 1) Our previous studies have identified a critical role of Delta4 (Dll4), an endothelial-specific membrane-bound ligand for Notch1 and Notch4, as a regulator of endothelial cell function. Dll4 is selectively expressed in the developing endothelium and is required for normal vascular development. Post-natally, Dll4 is expressed in the angiogenic endothelium, particularly in the tumor vasculature. We have found that Dll4 is a negative regulator of angiogenesis, as it functions as a selective inhibitor of VEGF-A by down-regulating the principal VEGF-A signaling receptor, VEGFR-2 and co-receptor neuropilin-1 (Npn-1). In pre-clinical cancer models, we have documented that Dll4 can markedly reduce tumor angiogenesis and the growth of tumors of lymphoid origin by reducing VEGF-A responses in the tumor vascular endothelium. In related experiments, we have begun to explore the role of the Notch ligand JAG2 in angiogenesis. To this end, we have developed a new mouse model of JAG2-deficiency and explored the potential contribution of Notch-dependent and Notch-independent pathways in endothelial cell function and angiogenesis. Preliminary results provide strong evidence for a role of Jag2 in the maintenance of the endothelial intestinal stem cell niche. 2) We have continued investigations on how ephrinB ligands and their EphB receptors orchestrate endothelial/endothelial/pericyte assembly in the vasculature. EphrinB ligands are surface-bound; thus receptor-ligand interactions in the B-type Eph/Ephrin interactions involve adjacent cells. In addition to activating their cognate EphB receptors, B-type Ephrins can function as signaling molecules when engaged by the receptor through reverse signaling. Eph receptors are tyrosine kinases interacting with their membrane-anchored ephrin ligands. We have now investigated the potential role of Eph/ephrin signaling in the regulation of endothelial cells survival. We have found that silencing EphrinB expression or expression of a tyrosine-phosphorylation-deficient mutant EphrinB (contains substitutions of all tyrosine residues that prevent tail phosphorylation and acts as a dominant-negative inhibitor of endogenous WT ephrin) causes endothelial cell death. This outcome cannot be prevented by the addition of exogenous VEGFA or FGF2. Biochemical and genetic experiments have revealed that such death is mediated by JNK3/MAPK10, and that EphrinB2 tyrosine phosphorylation-dependent signaling serves as a modulator of MAPK10/JNK3 expression. Thus, the silencing of JNK3 prevents cell death in endothelial cells that are EphrinB signaling-deficient. Consistent with these results, the hyaloid vasculature in mice genetically-deficient of EphrinB2 undergoes increased cell death in association with JNK3 activation, and JNK3-deficient mice display ocular vascular defects that mirror those of EphrinB2 signaling deficiency. These results provide evidence supporting a role for EphrinB signaling as an endothelial pro-survival pathway and a therapeutic target for inhibition of angiogenesis. Based on this evidence, we are currently exploring the possibility of targeting EphrinB2 signaling in the tumor vasculature to induce vessel regression and promote tumor cell starvation of collapse. 3) Pursuing this observation, we have explored different approaches to block EphrinB2-derived pro-survival signals in the vasculature. We have identified the SHP2 inhibitor SHP099 as a potent inhibitor of phospho-EphrinB2-STAT signaling and a selective inducer of endothelial cell death in vitro and in vivo. We have characterized the signaling consequences of SHP2 inhibition in endothelial cells and exploited this information to selectively target the tumor vasculature rather than the tumor cells. 4) In earlier observations we have linked the loss of the tumor-suppressor protein DLC1 with increased survival in primary endothelial cells under conditions of stress. We now discovered that DLC1 isa critical regulator of cell contact inhibition of proliferation in primary human endothelial cells, promoting cell death when the cells reach high density. DLC1 depletion confers a pro-survival phenotype to confluent, but not sparse endothelial cells, attributable to increased NF-kB activation associated with increased tumor necrosis factor alpha-induced protein 3 (TNFAIP3/A20) signaling. Consistent with a role of DLC1 depletion in endothelial cell tumorigenesis, we found that DLC1 is abnormally low and TNFAIP3/A20 is abnormally high in human angiosarcoma. Experimental treatment with the NF-kB inhibitor Tanespimycin/17-AAG significantly reduced angiosarcoma tumor growth in mouse. These results identify DLC1 as a previously unrecognized regulator of endothelial cell contact inhibition of proliferation that is depleted in angiosarcoma, and provide evidence supporting the targeting of NF-kB for the treatment of angiosarcoma where DLC1 is lost. 5) Based on the observation that the tumor suppressor DLC1 protein (our results) and the transcriptional co-activator YAP (literature) regulate cell-contact inhibition of growth, we have explored biochemical interactions between DLC1 and YAP. We found that DLC1 is a regulator of YAP and that the transcriptional co-activator function of YAP are required for the loss of cell-contact inhibition manifested by DLC1-deficient endothelial cells. If YAP is depleted from endothelial cells, DLC1-depleted endothelial cells stop growing when confluent and fail to pile-up. These results in vitro were corroborated by the observation that angiosarcoma tissues contain a significant proportion of DLC1-negative malignant endothelial cells where YAP is nuclear and active. This is not the case in the normal skin vasculature adjacent to the tumor. Verteporfin, an inhibitor of YAP, displayed a cleat ant-tumor effect in an experimental model of angiosarcoma in mice.